The primary focus of this project is on the development of improved statistical techniques for estimating potential human cancer risk using data generated from animal laboratory experiments. Current research efforts are particularly concerned with an assessment of the Gamma multi-hit model and an evaluation of the performance of multistage model in a megamouse experimental framework. The adaptation of methods for species-to-species extrapolation and the incorporation of pharmacokinetics in low-dose risk estimation also continue to be of interest.